Reading Comprehension on PaperQuiz
75.4Detail ScoreP2P
Evaluation results
| Method | Links | ||
|---|---|---|---|
| P2PGeneration Protocol=Poster Pipeline2026.07 | 75.4 | 95.6 | |
| Claude-4.8 OpusGeneration Protocol=Single-shot2026.07 | 66.67 | 93.37 | |
| GPT-5.5Generation Protocol=Single-shot2026.07 | 65.63 | 93.2 | |
| Paper2Poster ToolGeneration Protocol=Poster Pipeline2026.07 | 63 | 95.65 | |
| Gemini-3.1 ProGeneration Protocol=Single-shot2026.07 | 61.43 | 92.43 | |
| PosterGenGeneration Protocol=Poster Pipeline2026.07 | 57.45 | 91.85 | |
| ResearchStudio-ReelGeneration Protocol=Poster Pipeline, System=Claude Code2026.07 | 55.43 | 90.88 | |
| ResearchStudio-ReelGeneration Protocol=Poster Pipeline, System=Codex2026.07 | 53.43 | 91.32 | |
| Author ground-truth2026.07 | 50.11 | 88.92 |